This notebook contains a set of analyses for analyzing WatchItPlayed’s boardgamegeek collection. The bulk of the analysis is focused on building a user-specific predictive model to predict the games that the specified user is likely to own. This enables us to ask questions like, based on the games the user currently owns, what games are a good fit for their collection? What upcoming games are they likely to purchase?
We can look at a basic description of the number of games that the user owns, has rated, has previously owned, etc.
What years has the user owned/rated games from? While we can’t see when a user added or removed a game from their collection, we can look at their collection by the years in which their games were published.
We can look at the most frequent types of categories, mechanics, designers, and artists that appear in a user’s collection.
We’ll examine predictive models trained on a user’s collection for games published through 2020. How many games has the user owned/rated/played in the training set (games prior to 2020)?
username | dataset | period | games_owned | games_rated |
WatchItPlayed | training | published before 2020 | 273 | 0 |
WatchItPlayed | validation | published 2020 | 45 | 0 |
WatchItPlayed | test | published after 2020 | 76 | 0 |
The main outcome we will be modeling for the user is owned, which refers to whether the user currently owns or has a previously owned a game in their collection. Our goal is to train a predictive model to learn the probability that a user will add a game to their collection based on its observable features. This amounts to looking at historical data and looking to find patterns that exist between features of games and games present in the user’s collection.
One of the models we trained was a decision tree, which looks for decision rules that can be used to separate games the user owns from games they don’t. The resulting model produces a decision corresponding to yes or no statements: to explain why the model predicts the user to own game, we start at the top of the tree and follow the rules that were learned from the training data.
Note: the tree below has been further pruned to make it easier to visualize.
Decision trees are highly interpretible models that are easy to train and can identify important interactions and nonlinearities present in the data. Individual trees have the drawback of being less predictive than other common models, but it can be useful to look at them to gain some understanding of key predictors and relationships found in the training data.
We can examine coefficients from another model we trained, which is a logistic regression with elastic net regularization (which I will refer to as a penalized logistic regression). Positive values indicate that a feature increases a user’s probability of owning/rating a game, while negative values indicate a feature decreases the probability. To be precise, the coefficients indicate the effect of a particular feature on the log-odds of a user owning a game.
Why did the model identify these features? We can make density plots of the important features for predicting whether the user owned a game. Blue indicates the density for games owned by the user, while grey indicates the density for games not owned by the user.
Binary predictors can be difficult to see with this visualization, so we can also directly examine the percentage of games in a user’s collection with a predictor vs the percentage of all games with that predictor.
% of Games with Feature | ||||
username | Feature | User_Collection | All_Games | Ratio |
WatchItPlayed | Asmodee | 15.8% | 2.5% | 6.35 |
WatchItPlayed | Stronghold Games | 3.3% | 0.6% | 5.83 |
WatchItPlayed | Fantasy Flight Games | 5.9% | 1.1% | 5.23 |
WatchItPlayed | Solo Solitaire Game | 15.4% | 3.2% | 4.87 |
WatchItPlayed | ZMan Games | 6.6% | 1.4% | 4.76 |
WatchItPlayed | GMT Games | 5.9% | 1.3% | 4.60 |
WatchItPlayed | Eagle-Gryphon Games | 2.6% | 0.7% | 3.42 |
WatchItPlayed | Combinatorial | 4.4% | 1.6% | 2.71 |
WatchItPlayed | 3D Components | 3.7% | 1.4% | 2.54 |
WatchItPlayed | Paper And Pencil | 5.9% | 2.3% | 2.53 |
WatchItPlayed | Deduction Game | 11.0% | 5.0% | 2.19 |
WatchItPlayed | Hand Management | 38.8% | 19.9% | 1.95 |
WatchItPlayed | Party Game | 17.6% | 9.3% | 1.90 |
WatchItPlayed | Dice Rolling | 24.2% | 28.5% | 0.85 |
WatchItPlayed | Trading | 1.1% | 3.0% | 0.36 |
WatchItPlayed | Pickup And Deliver | 0.4% | 2.7% | 0.13 |
Before predicting games in upcoming years, we can examine how well the model did and what games it liked in the training set. In this case, we used resampling techniques (cross validation) to ensure that the model had not seen a game before making its predictions.
Displaying the 100 games from the training set with the highest probability of ownership, highlighting in blue games the user has owned.
Rank | Published | ID | Name | Pr(Owned) | Owned |
1 | 2019 | 286096 | Tapestry | 0.957 | yes |
2 | 2016 | 169786 | Scythe | 0.880 | yes |
3 | 2018 | 205896 | Rising Sun | 0.859 | no |
4 | 2013 | 143693 | Glass Road | 0.823 | no |
5 | 2017 | 220308 | Gaia Project | 0.748 | no |
6 | 2019 | 281946 | Aftermath | 0.719 | no |
7 | 2010 | 62227 | Labyrinth: The War on Terror, 2001 – ? | 0.715 | no |
8 | 2017 | 233078 | Twilight Imperium: Fourth Edition | 0.715 | no |
9 | 2018 | 244711 | Newton | 0.688 | no |
10 | 2010 | 70512 | Luna | 0.615 | no |
11 | 2016 | 167791 | Terraforming Mars | 0.610 | yes |
12 | 2017 | 193728 | Pendragon: The Fall of Roman Britain | 0.589 | yes |
13 | 2018 | 260428 | Pandemic: Fall of Rome | 0.544 | no |
14 | 2019 | 266192 | Wingspan | 0.542 | yes |
15 | 2012 | 123096 | Space Cadets | 0.540 | no |
16 | 2010 | 65200 | Asteroyds | 0.526 | no |
17 | 2010 | 62219 | Dominant Species | 0.519 | no |
18 | 2019 | 220558 | Ancient Civilizations of the Inner Sea | 0.504 | no |
19 | 2019 | 270971 | Era: Medieval Age | 0.487 | no |
20 | 2016 | 176083 | Hit Z Road | 0.482 | no |
21 | 2019 | 265736 | Tiny Towns | 0.480 | yes |
22 | 2016 | 205637 | Arkham Horror: The Card Game | 0.461 | yes |
23 | 2017 | 174430 | Gloomhaven | 0.459 | no |
24 | 2014 | 159508 | AquaSphere | 0.456 | yes |
25 | 2012 | 124742 | Android: Netrunner | 0.423 | no |
26 | 2014 | 145371 | Three Kingdoms Redux | 0.419 | no |
27 | 2016 | 187645 | Star Wars: Rebellion | 0.403 | yes |
28 | 1982 | 2511 | Sherlock Holmes Consulting Detective: The Thames Murders & Other Cases | 0.399 | no |
29 | 2017 | 199904 | Pericles: The Peloponnesian Wars | 0.394 | no |
30 | 2018 | 245654 | Railroad Ink: Deep Blue Edition | 0.388 | no |
31 | 2018 | 199792 | Everdell | 0.373 | no |
32 | 2015 | 183394 | Viticulture Essential Edition | 0.360 | no |
33 | 2006 | 21882 | Blue Moon City | 0.352 | no |
34 | 2017 | 226320 | My Little Scythe | 0.344 | no |
35 | 2014 | 159675 | Fields of Arle | 0.343 | yes |
36 | 2017 | 230059 | Crossfire | 0.343 | no |
37 | 2014 | 163412 | Patchwork | 0.340 | yes |
38 | 2015 | 181304 | Mysterium | 0.336 | no |
39 | 2006 | 22141 | Cleopatra and the Society of Architects | 0.330 | no |
40 | 2019 | 285774 | Marvel Champions: The Card Game | 0.319 | no |
41 | 2016 | 163154 | Falling Sky: The Gallic Revolt Against Caesar | 0.317 | yes |
42 | 2011 | 70919 | Takenoko | 0.315 | yes |
43 | 2010 | 73439 | Troyes | 0.311 | no |
44 | 2015 | 175878 | 504 | 0.311 | no |
45 | 2015 | 170216 | Blood Rage | 0.306 | yes |
46 | 2019 | 276025 | Maracaibo | 0.302 | no |
47 | 2017 | 162886 | Spirit Island | 0.301 | no |
48 | 2013 | 133848 | Euphoria: Build a Better Dystopia | 0.300 | no |
49 | 2006 | 25417 | BattleLore | 0.298 | no |
50 | 2012 | 129622 | Love Letter | 0.298 | no |
51 | 2017 | 197376 | Charterstone | 0.296 | no |
52 | 2004 | 10630 | Memoir '44 | 0.295 | no |
53 | 2011 | 103343 | A Game of Thrones: The Board Game (Second Edition) | 0.290 | no |
54 | 2016 | 160010 | Conan | 0.285 | no |
55 | 2017 | 234671 | Pandemic: Rising Tide | 0.281 | no |
56 | 2005 | 15062 | Shadows over Camelot | 0.269 | no |
57 | 2015 | 161226 | The Builders: Antiquity | 0.267 | no |
58 | 2016 | 205158 | Codenames: Deep Undercover | 0.266 | no |
59 | 2011 | 84419 | Space Empires 4X | 0.263 | no |
60 | 2009 | 40692 | Small World | 0.263 | yes |
61 | 2017 | 209778 | Magic Maze | 0.263 | no |
62 | 2016 | 193037 | Dead of Winter: The Long Night | 0.262 | no |
63 | 2011 | 102548 | Dungeon Fighter | 0.258 | no |
64 | 2019 | 285984 | Last Bastion | 0.255 | no |
65 | 2017 | 221805 | Breaking Bad: The Board Game | 0.253 | no |
66 | 2012 | 104162 | Descent: Journeys in the Dark (Second Edition) | 0.251 | no |
67 | 2011 | 69552 | Panic Station | 0.233 | no |
68 | 2015 | 168435 | Between Two Cities | 0.232 | no |
69 | 2018 | 256916 | Concordia Venus | 0.231 | no |
70 | 1999 | 88 | Torres | 0.227 | no |
71 | 2019 | 253344 | Cthulhu: Death May Die | 0.223 | no |
72 | 2013 | 133528 | City of Remnants | 0.222 | no |
73 | 2014 | 157354 | Five Tribes | 0.222 | no |
74 | 2018 | 246297 | Shadows: Amsterdam | 0.222 | no |
75 | 2013 | 144239 | Impulse | 0.220 | no |
76 | 2016 | 177736 | A Feast for Odin | 0.219 | no |
77 | 2018 | 233080 | Book of Dragons | 0.217 | no |
78 | 2009 | 43868 | The Adventurers: The Temple of Chac | 0.216 | no |
79 | 2012 | 121921 | Robinson Crusoe: Adventures on the Cursed Island | 0.213 | no |
80 | 2016 | 200680 | Agricola (Revised Edition) | 0.213 | no |
81 | 2015 | 182694 | Watson & Holmes | 0.212 | no |
82 | 2017 | 164949 | Time of Crisis: The Roman Empire in Turmoil, 235-284 AD | 0.209 | yes |
83 | 2018 | 240464 | Cosmic Run: Regeneration | 0.209 | no |
84 | 2019 | 259081 | Machi Koro Legacy | 0.205 | no |
85 | 2004 | 9220 | Saboteur | 0.203 | no |
86 | 2017 | 232988 | The Castles of Burgundy: The Dice Game | 0.203 | no |
87 | 2019 | 272453 | KeyForge: Age of Ascension | 0.201 | no |
88 | 2014 | 158899 | Colt Express | 0.201 | no |
89 | 2014 | 148228 | Splendor | 0.201 | yes |
90 | 2016 | 155821 | Inis | 0.195 | yes |
91 | 1997 | 42 | Tigris & Euphrates | 0.194 | no |
92 | 1971 | 594 | Sleuth | 0.194 | no |
93 | 2014 | 161547 | Timeline: Americana | 0.193 | no |
94 | 2018 | 257321 | Gen7: A Crossroads Game | 0.192 | no |
95 | 2015 | 163968 | Elysium | 0.192 | yes |
96 | 2017 | 180199 | Colonial Twilight: The French-Algerian War, 1954-62 | 0.192 | yes |
97 | 2009 | 54043 | Jaipur | 0.191 | yes |
98 | 2016 | 195856 | Bloodborne: The Card Game | 0.191 | no |
99 | 2014 | 146886 | La Granja | 0.191 | no |
100 | 2014 | 154203 | Imperial Settlers | 0.190 | no |
This section contains a variety of visualizations and metrics for assessing the performance of the model(s) during resampling. If you’re not particularly interested in predictive modeling, skip down further to the predictions from the model.
An easy way to examine the performance of classification model is to view a separation plot. We plot the predicted probabilities from the model for every game (from resampling) from lowest to highest. We then overlay a blue line for any game that the user does own. A good classifier is one that is able to separate the blue (games owned by the user) from the white (games not owned by the user), with most of the blue occurring at the highest probabilities (right side of the chart).
We can more formally assess how well each model did in resampling by looking at the area under the receiver operating characteristic curve. A perfect model would receive a score of 1, while a model that cannot predict the outcome will default to a score of 0.5. The extent to which something is a good score depends on the setting, but generally anything in the .8 to .9 range is very good while the .7 to .8 range is perfectly acceptable.
wflow_id | .metric | .estimator | .estimate |
GLM | roc_auc | binary | 0.85 |
Decision Tree | roc_auc | binary | 0.75 |
Another way to think about the model performance is to view its lift, or its ability to detect the positive outcomes over that of a null model. High lift indicates the model can much more quickly find all of the positive outcomes (in this case, games owned or played by the user), while a model with no lift is no better than random guessing. A gains chart is another way to view this.
While we are probably more interested in the lift provided by the models to evaluate their efficacy, we can also explore the optimal cutpoint if we wanted to define a hard threshold for identifying games a user will own vs not own.
The threshold we select depends on how we much we care about false positives (games the model predicts that the user does not own) vs false negatives (games the user owns that the model does not predict). We can toggle threshold to
Finally, we can understand the performance of the model by examining its calibration. If the model assigns a probability of 5%, how often does the outcome actually occur? A well calibrated model is one in which the predicted probabilities reflect the probabilities we would observe in the actual data. We can assess the calibration of a model by grouping its predictions into bins and assessing how often we observe the outcome versus how often our model expects to observe the outcome.
A model that is well calibrated will closely follow the dashed line - its expected probabilities match that of the observed probabilities. A model that consistently underestimates the probability of the event will be over this dashed line, be a while a model that overestimates the probability will be under the dashed line.
What games does the model think WatchItPlayed is most likely to own that are not in their collection?
Published | ID | Name | Pr(Owned) | Owned |
2018 | 205896 | Rising Sun | 0.859 | no |
2013 | 143693 | Glass Road | 0.823 | no |
2017 | 220308 | Gaia Project | 0.748 | no |
2019 | 281946 | Aftermath | 0.719 | no |
2010 | 62227 | Labyrinth: The War on Terror, 2001 – ? | 0.715 | no |
What games does the model think WatchItPlayed is least likely to own that are in their collection?
Published | ID | Name | Pr(Owned) | Owned |
2003 | 12171 | Le Passe-Trappe | 0.003 | yes |
2013 | 139747 | Arena Rex | 0.004 | yes |
1972 | 4079 | What's My Word? | 0.004 | yes |
1960 | 148203 | Dutch Blitz | 0.004 | yes |
1971 | 2223 | UNO | 0.005 | yes |
Top 25 games most likely to be owned by the user in each year, highlighting in blue the games that the user has owned.
rank | 2012 | 2013 | 2014 | 2015 | 2016 | 2017 | 2018 | 2019 |
1 | Space Cadets | Glass Road | AquaSphere | Viticulture Essential Edition | Scythe | Gaia Project | Rising Sun | Tapestry |
2 | Android: Netrunner | Euphoria: Build a Better Dystopia | Three Kingdoms Redux | Mysterium | Terraforming Mars | Twilight Imperium: Fourth Edition | Newton | Aftermath |
3 | Love Letter | City of Remnants | Fields of Arle | 504 | Hit Z Road | Pendragon: The Fall of Roman Britain | Pandemic: Fall of Rome | Wingspan |
4 | Descent: Journeys in the Dark (Second Edition) | Impulse | Patchwork | Blood Rage | Arkham Horror: The Card Game | Gloomhaven | Railroad Ink: Deep Blue Edition | Ancient Civilizations of the Inner Sea |
5 | Robinson Crusoe: Adventures on the Cursed Island | Timeline: Music & Cinema | Five Tribes | The Builders: Antiquity | Star Wars: Rebellion | Pericles: The Peloponnesian Wars | Everdell | Era: Medieval Age |
6 | Terra Mystica | Cappuccino | Colt Express | Between Two Cities | Falling Sky: The Gallic Revolt Against Caesar | My Little Scythe | Concordia Venus | Tiny Towns |
7 | Escape: The Curse of the Temple | The Builders: Middle Ages | Splendor | Watson & Holmes | Conan | Crossfire | Shadows: Amsterdam | Marvel Champions: The Card Game |
8 | Zombicide | A Distant Plain | Timeline: Americana | Elysium | Codenames: Deep Undercover | Spirit Island | Book of Dragons | Maracaibo |
9 | Mafia: Vendetta | The Hunters: German U-Boats at War, 1939-43 | La Granja | The Game | Dead of Winter: The Long Night | Charterstone | Cosmic Run: Regeneration | Last Bastion |
10 | Kemet | Room 25 | Imperial Settlers | Pirates of the 7 Seas | A Feast for Odin | Pandemic: Rising Tide | Gen7: A Crossroads Game | Cthulhu: Death May Die |
11 | Agricola: All Creatures Big and Small | Crossing | Black Fleet | Runebound (Third Edition) | Agricola (Revised Edition) | Magic Maze | Kero | Machi Koro Legacy |
12 | Mice and Mystics | Heads Up!: Party Game | Camel Up | Star Trek: Five-Year Mission | Inis | Breaking Bad: The Board Game | Carpe Diem | KeyForge: Age of Ascension |
13 | Andean Abyss | BANG! The Dice Game | King of New York | Mombasa | Bloodborne: The Card Game | Time of Crisis: The Roman Empire in Turmoil, 235-284 AD | Lords of Hellas | Caylus 1303 |
14 | Yedo | Tash-Kalar: Arena of Legends | Nations: The Dice Game | T.I.M.E Stories | Cottage Garden | The Castles of Burgundy: The Dice Game | The World of SMOG: Rise of Moloch | Silver & Gold |
15 | Libertalia | Cardline: Globetrotter | Deception: Murder in Hong Kong | The King Is Dead | Encore! | Colonial Twilight: The French-Algerian War, 1954-62 | SpaceCorp: 2025-2300AD | Pandemic: Rapid Response |
16 | Timeline | Legacy: The Testament of Duke de Crecy | Medina (Second Edition) | Timeline: Star Wars | Citadels | Richard the Lionheart | Hitler's Reich: WW2 in Europe | Gandhi: The Decolonization of British India, 1917 – 1947 |
17 | Timeline: General Interest | Viticulture | Fire in the Lake | Timeline: Québec | Liberty or Death: The American Insurrection | Secrets | Between Two Castles of Mad King Ludwig | The Magnificent |
18 | Cardline: Animals | BioShock Infinite: The Siege of Columbia | Onirim (Second Edition) | My Village | Aeon's End | Sherlock Holmes Consulting Detective: Vanishing from Hyde Park | Forum Trajanum | Herbaceous Sprouts |
19 | The Resistance: Avalon | Navajo Wars | Emperor's New Clothes | Mission: Red Planet (Second Edition) | Great Western Trail | Dinosaur Island | Fort Sumter: The Secession Crisis, 1860-61 | The Mind Extreme |
20 | I'm the Boss!: The Card Game | Lewis & Clark: The Expedition | DungeonQuest Revised Edition | Triumph & Tragedy: European Balance of Power 1936-1945 | Coal Baron: The Great Card Game | Unlock!: Escape Adventures – Doo-Arann Dungeon | Timeline: Science Museum | Tainted Grail: The Fall of Avalon |
21 | 1989: Dawn of Freedom | Concordia | La Isla | The Voyages of Marco Polo | The Oracle of Delphi | LYNGK | Timeline: Classic | Sierra West |
22 | Coup | Rococo | Pandemic: The Cure | Sylvion | Sherlock Holmes Consulting Detective: Jack the Ripper & West End Adventures | Warhammer Underworlds: Shadespire | Brikks | Unmatched: Battle of Legends, Volume One |
23 | Divinare | Time 'n' Space | Star Wars: Imperial Assault | Cardline: Animals 2 | Pandemic: Reign of Cthulhu | The Godfather: Corleone's Empire | Root | Unmatched Game System |
24 | Noah | Cuba Libre | Kill Shakespeare | Unusual Suspects | Codenames: Pictures | Summit: The Board Game | Arraial | Mental Blocks |
25 | Il Vecchio | Caverna: The Cave Farmers | Gaïa | Pandemic Legacy: Season 1 | Game of Thrones: The Iron Throne | SpyNet | Warhammer Underworlds: Nightvault | The Only Word: the Party Word Game |
This is an interactive table for the model’s predictions for the training set (from resampling).
We’ll validate the model by looking at its predictions for games published in 2020. That is, how well did a model trained on a user’s collection through 2020 perform in predicting games for the user in 2020?
username | outcome | dataset | method | .metric | .estimate |
WatchItPlayed | owned | validation | GLM | roc_auc | 0.728 |
WatchItPlayed | owned | validation | Decision Tree | roc_auc | 0.671 |
Table of top 50 games from 2020, highlighting games that the user owns.
Published | ID | Name | Pr(Owned) | Owned |
2020 | 312804 | Pendulum | 0.432 | no |
2020 | 291457 | Gloomhaven: Jaws of the Lion | 0.363 | yes |
2020 | 314040 | Pandemic Legacy: Season 0 | 0.355 | yes |
2020 | 306481 | Tawantinsuyu: The Inca Empire | 0.346 | no |
2020 | 184267 | On Mars | 0.333 | no |
2020 | 304420 | Bonfire | 0.332 | no |
2020 | 253506 | Versailles 1919 | 0.305 | no |
2020 | 256317 | Guild Master | 0.238 | no |
2020 | 292333 | Cowboys II: Cowboys & Indians Edition | 0.218 | no |
2020 | 300322 | Hallertau | 0.209 | no |
2020 | 296626 | Sonora | 0.199 | no |
2020 | 296151 | Viscounts of the West Kingdom | 0.180 | yes |
2020 | 319966 | The King Is Dead: Second Edition | 0.167 | yes |
2020 | 189664 | The Hunted: Twilight of the U-Boats, 1943-45 | 0.161 | no |
2020 | 246900 | Eclipse: Second Dawn for the Galaxy | 0.160 | no |
2020 | 294484 | Unmatched: Cobble & Fog | 0.159 | yes |
2020 | 315060 | Unmatched: Buffy the Vampire Slayer | 0.159 | yes |
2020 | 299179 | Chancellorsville 1863 | 0.144 | no |
2020 | 316554 | Dune: Imperium | 0.131 | yes |
2020 | 298371 | Wild Space | 0.131 | no |
2020 | 301880 | Raiders of Scythia | 0.129 | no |
2020 | 325635 | Unmatched: Little Red Riding Hood vs. Beowulf | 0.128 | yes |
2020 | 298572 | Cosmic Encounter Duel | 0.120 | no |
2020 | 284777 | Unmatched: Jurassic Park – InGen vs Raptors | 0.118 | no |
2020 | 255456 | Beneath the Med: Regia Marina at Sea 1940-1943 | 0.117 | no |
2020 | 316412 | The LOOP | 0.112 | no |
2020 | 316377 | 7 Wonders (Second Edition) | 0.111 | no |
2020 | 301767 | Mysterium Park | 0.110 | yes |
2020 | 296345 | Sherlock Holmes Consulting Detective: The Baker Street Irregulars | 0.108 | no |
2020 | 304285 | Infinity Gauntlet: A Love Letter Game | 0.105 | no |
2020 | 302425 | Unlock!: Mythic Adventures | 0.105 | no |
2020 | 294788 | Conqueror: Final Conquest | 0.101 | no |
2020 | 229782 | Roland Wright: The Dice Game | 0.100 | no |
2020 | 274841 | Cóatl | 0.099 | no |
2020 | 303672 | Trek 12: Himalaya | 0.097 | no |
2020 | 318183 | Prehistories | 0.097 | no |
2020 | 312267 | Star Wars: Unlock! | 0.097 | no |
2020 | 301919 | Pandemic: Hot Zone – North America | 0.094 | yes |
2020 | 319794 | Timeline: Chile | 0.093 | no |
2020 | 282081 | The Zorro Dice Game | 0.093 | no |
2020 | 233262 | Tidal Blades: Heroes of the Reef | 0.093 | no |
2020 | 318084 | Furnace | 0.092 | yes |
2020 | 296512 | The Game: Quick & Easy | 0.091 | no |
2020 | 297030 | Tekhenu: Obelisk of the Sun | 0.091 | no |
2020 | 302465 | Obsidia | 0.090 | no |
2020 | 281075 | Welcome to New Las Vegas | 0.085 | no |
2020 | 252153 | Tang Garden | 0.083 | no |
2020 | 242520 | All Bridges Burning: Red Revolt and White Guard in Finland, 1917-1918 | 0.081 | yes |
2020 | 303669 | Magic Rabbit | 0.080 | no |
2020 | 313698 | Monster Expedition | 0.079 | no |
We can then refit our model to the training and validation set in order to predict all upcoming games for the user.
Examine the top 100 upcoming games, highlighting in blue ones the user already owns.
Rank | Published | ID | Name | Pr(Owned) | Owned |
1 | 2022 | 310873 | Carnegie | 0.625 | no |
2 | 2022 | 349067 | The Lord of the Rings: The Card Game – Revised Core Set | 0.482 | no |
3 | 2023 | 349793 | Age of Rome | 0.451 | no |
4 | 2021 | 285967 | Ankh: Gods of Egypt | 0.450 | no |
5 | 2022 | 256680 | Return to Dark Tower | 0.390 | yes |
6 | 2021 | 329465 | Red Rising | 0.352 | no |
7 | 2021 | 342073 | Berried Treasure | 0.348 | no |
8 | 2022 | 317511 | Tindaya | 0.320 | no |
9 | 2023 | 347909 | Rogue Angels: Legacy of the Burning Suns | 0.318 | no |
10 | 2021 | 305682 | Rolling Realms | 0.290 | yes |
11 | 2022 | 335764 | Unmatched: Battle of Legends, Volume Two | 0.269 | yes |
12 | 2022 | 326933 | Unmatched: Redemption Row | 0.262 | yes |
13 | 2022 | 326934 | Unmatched: Hell's Kitchen | 0.262 | yes |
14 | 2022 | 315610 | Massive Darkness 2: Hellscape | 0.257 | no |
15 | 2021 | 340466 | Unfathomable | 0.252 | no |
16 | 2021 | 304783 | Hadrian's Wall | 0.243 | yes |
17 | 2022 | 295770 | Frosthaven | 0.239 | no |
18 | 2022 | 356033 | Libertalia: Winds of Galecrest | 0.238 | yes |
19 | 2022 | 341945 | La Granja: Deluxe Master Set | 0.228 | no |
20 | 2021 | 339789 | Welcome to the Moon | 0.224 | no |
21 | 2021 | 249277 | Brazil: Imperial | 0.224 | no |
22 | 2021 | 298102 | Roll Camera!: The Filmmaking Board Game | 0.221 | no |
23 | 2021 | 340237 | Wonder Book | 0.219 | no |
24 | 2022 | 331106 | The Witcher: Old World | 0.219 | no |
25 | 2021 | 343905 | Boonlake | 0.218 | yes |
26 | 2022 | 284778 | Unmatched: Jurassic Park – Dr. Sattler vs. T. Rex | 0.208 | yes |
27 | 2021 | 339906 | The Hunger | 0.180 | no |
28 | 2021 | 342942 | Ark Nova | 0.175 | no |
29 | 2021 | 273330 | Bloodborne: The Board Game | 0.173 | no |
30 | 2021 | 262941 | Dominant Species: Marine | 0.169 | yes |
31 | 2022 | 266064 | Trudvang Legends | 0.163 | no |
32 | 2021 | 295947 | Cascadia | 0.160 | yes |
33 | 2021 | 344277 | Corrosion | 0.156 | no |
34 | 2022 | 305096 | Endless Winter: Paleoamericans | 0.154 | no |
35 | 2021 | 296577 | Red Flag Over Paris | 0.148 | yes |
36 | 2021 | 306202 | Philosophia: Floating World | 0.146 | no |
37 | 2021 | 313730 | Harsh Shadows | 0.142 | no |
38 | 2021 | 314491 | Meadow | 0.134 | no |
39 | 2021 | 290236 | Canvas | 0.132 | no |
40 | 2021 | 332944 | Sobek: 2 Players | 0.128 | no |
41 | 2021 | 344258 | That Time You Killed Me | 0.126 | no |
42 | 2022 | 352263 | Through Ice and Snow | 0.126 | no |
43 | 2021 | 286751 | Zombicide: 2nd Edition | 0.124 | yes |
44 | 2021 | 339790 | Cocktail | 0.124 | no |
45 | 2021 | 338834 | MicroMacro: Crime City – Full House | 0.124 | no |
46 | 2021 | 308989 | Bristol 1350 | 0.121 | no |
47 | 2021 | 304985 | Dark Ages: Holy Roman Empire | 0.120 | no |
48 | 2021 | 295535 | Dark Ages: Heritage of Charlemagne | 0.120 | no |
49 | 2021 | 338980 | Eastern Empires | 0.119 | no |
50 | 2021 | 336382 | Marvel United: X-Men | 0.118 | no |
51 | 2021 | 340909 | Gloomholdin' | 0.118 | no |
52 | 2021 | 333553 | For the King (and Me) | 0.116 | no |
53 | 2021 | 316287 | Quest | 0.115 | no |
54 | 2023 | 312959 | Rallyman: DIRT | 0.114 | no |
55 | 2022 | 350316 | Wayfarers of the South Tigris | 0.114 | no |
56 | 2022 | 280726 | Legacies | 0.112 | no |
57 | 2022 | 322524 | Bardsung | 0.111 | no |
58 | 2021 | 342848 | World of Warcraft: Wrath of the Lich King | 0.109 | yes |
59 | 2021 | 341169 | Great Western Trail (Second Edition) | 0.108 | no |
60 | 2021 | 291572 | Oath: Chronicles of Empire and Exile | 0.108 | no |
61 | 2021 | 331635 | Kameloot | 0.105 | no |
62 | 2021 | 299255 | Vienna Connection | 0.105 | yes |
63 | 2022 | 347703 | First Rat | 0.104 | no |
64 | 2021 | 345435 | Fireball Island: Race to Adventure | 0.103 | no |
65 | 2021 | 340677 | Bad Company | 0.102 | no |
66 | 2022 | 240980 | Blood on the Clocktower | 0.100 | no |
67 | 2021 | 301366 | Caves of Rwenzori | 0.096 | no |
68 | 2021 | 281248 | Cape May | 0.095 | no |
69 | 2022 | 353470 | Star Wars: Jabba's Palace – A Love Letter Game | 0.095 | no |
70 | 2021 | 283242 | The Whatnot Cabinet | 0.094 | no |
71 | 2021 | 331549 | MiniQuest Adventures | 0.093 | no |
72 | 2022 | 338460 | The Isle of Cats: Explore & Draw | 0.092 | no |
73 | 2022 | 319807 | Shogun no Katana | 0.091 | no |
74 | 2021 | 307862 | Dollars to Donuts | 0.089 | no |
75 | 2021 | 329670 | Pandemic: Hot Zone – Europe | 0.089 | yes |
76 | 2021 | 318184 | Imperium: Classics | 0.089 | no |
77 | 2022 | 299106 | Fractal: Beyond the Void | 0.088 | no |
78 | 2021 | 292375 | The Great Wall | 0.086 | no |
79 | 2021 | 319263 | One Card Dungeon | 0.086 | no |
80 | 2021 | 319899 | Decktective: Nightmare in the Mirror | 0.084 | no |
81 | 2021 | 320446 | Corduba 27 a.C. | 0.083 | no |
82 | 2022 | 273814 | Deliverance | 0.083 | no |
83 | 2021 | 251747 | Atlantic Chase | 0.082 | no |
84 | 2022 | 320718 | Hidden Leaders | 0.080 | no |
85 | 2021 | 318322 | Hidden Strike: American Revolution | 0.080 | no |
86 | 2021 | 259066 | Commands & Colors: Samurai Battles | 0.080 | no |
87 | 2022 | 334065 | Verdant | 0.079 | no |
88 | 2021 | 324242 | Sheepy Time | 0.078 | no |
89 | 2022 | 346199 | A Game of Thrones: B'Twixt | 0.076 | no |
90 | 2021 | 283387 | Rocketmen | 0.076 | no |
91 | 2021 | 299566 | Batman: The Animated Series Adventures – Shadow of the Bat | 0.075 | no |
92 | 2022 | 342444 | Black Rose Wars: Rebirth | 0.075 | no |
93 | 2022 | 338067 | 6: Siege – The Board Game | 0.075 | no |
94 | 2021 | 291847 | Mantis Falls | 0.073 | no |
95 | 2021 | 260524 | Beyond Humanity: Colonies | 0.072 | no |
96 | 2022 | 295374 | Long Shot: The Dice Game | 0.071 | no |
97 | 2021 | 320960 | Roll In One | 0.071 | no |
98 | 2021 | 259394 | Storm Above the Reich | 0.071 | no |
99 | 2021 | 343696 | Dune: Betrayal | 0.071 | no |
100 | 2021 | 262201 | Sword & Sorcery: Ancient Chronicles | 0.071 | no |
## used (Mb) gc trigger (Mb) limit (Mb) max used (Mb)
## Ncells 6037822 322.5 13980205 746.7 NA 13980205 746.7
## Vcells 172512110 1316.2 558901608 4264.1 102400 1113448843 8495.0